FedDC: Federated Learning with Non-IID Data via Local Drift Decoupling and Correction

TL;DR

FedDC is a novel federated learning algorithm that effectively addresses data heterogeneity by decoupling and correcting local model drift, leading to faster convergence and improved performance across diverse non-iid data scenarios.

Contribution

Introduces FedDC, a lightweight method with local drift correction that enhances federated learning robustness and efficiency with non-iid data.

Findings

Faster convergence in federated learning tasks.

Improved accuracy on image classification with non-iid data.

Robust performance under partial participation and heterogeneous clients.

Abstract

Federated learning (FL) allows multiple clients to collectively train a high-performance global model without sharing their private data. However, the key challenge in federated learning is that the clients have significant statistical heterogeneity among their local data distributions, which would cause inconsistent optimized local models on the client-side. To address this fundamental dilemma, we propose a novel federated learning algorithm with local drift decoupling and correction (FedDC). Our FedDC only introduces lightweight modifications in the local training phase, in which each client utilizes an auxiliary local drift variable to track the gap between the local model parameter and the global model parameters. The key idea of FedDC is to utilize this learned local drift variable to bridge the gap, i.e., conducting consistency in parameter-level. The experiment results and analysis demonstrate that FedDC yields expediting convergence and better performance on various image classification tasks, robust in partial participation settings, non-iid data, and heterogeneous clients.